Let’s Go To The Movies

oscars <- oscars %>%    # rename year for join
  rename('year' = 'year_film')

meta$year <-  format(as.Date(meta$release_date, format = "%m/%d/%Y"), "%Y")

blue = '#000080' # just the color code I wanted to use for my print
  
meta <- meta %>% 
  mutate(year = as.numeric(year)) # easier use for comparisons/inflation

# renames
budget <- budget %>% 
  rename(vote_average = score) %>% # renaming vote data for joins etc
  rename(vote_count = votes) %>% 
  rename(Title = 'Movie Title')


oscars <- oscars %>% 
  rename(Title = 'film')  # rename for joins

adj <- inflation %>% 
  mutate(multiplier = (22.82/amount)) 
# create multiplier column for easy calculations

budget <- budget %>% 
  rename(new_budget = Budget)  # rename for joins etc


options(scipen = 100) # avoid scientific notation 

full_budget <- full_join(meta, budget, on = 'Title') 

# full join of metadata and budget data to get new and old movies etc

full_bud <- full_budget %>% 
  mutate(budget = pmax(new_budget, meta_budget, na.rm = T)) %>% 
  select(Title,genres, budget,new_budget,
         meta_budget, popularity,year,
         release_date, revenue, runtime, vote_average,
         vote_count,gross)
# set budget to max of two different budgets.
# picking max is arbitrary, but needed in most cases


full_bud <- full_bud[-c(1,2,3),] %>%
   arrange(desc(as.numeric(budget)))
# remove first three unnecessary rows


full_bud <- full_bud %>%  # replace 1900 sentinels with 2022
  mutate(year = replace(year, year == 1900, 2022))
# Most of these seemed to be less known movies
# so I thought that 2022 would do the least harm 
# with the inflation numbers

adj_bud <- full_join(full_bud, adj,
                             on = c('Title', 'year')) %>% 
  mutate(with_inflation = (as.numeric(budget) * (multiplier))) %>% 
  mutate(gross_inflation = (as.numeric(full_bud$gross)
                            * (multiplier))) %>% 
  select(Title,genres,vote_average, vote_count, budget,
         with_inflation, gross_inflation, year, gross,
         release_date) %>% 
  arrange(desc(with_inflation))  # joining movies with inflation

# creating with_inflation(budget) and gross_inflation columns
# arranging by highest with_inflation budgets


osc_bud <-  full_join(adj_bud, oscars, on  = c('Title'))

# joining all other data to Oscars data
# This is the starting point for most of my data manipulation
# 62706 entries, 15 total columns, of which 8 columns used.